Factual error rate on difficult prompts · Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified. · 测试定义与成绩

Factual error rate on difficult prompts · Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified. 收录 5 次模型发布的评测披露或成绩。

测试版本与指标

别名 Factual error rate on difficult prompts (lower is better)

  • Responses with at least one factual error · % · 越低越好

各系列最新模型 · 公布成绩

此测试使用完整公开配置。排名按指标与评测协议分组,已记录的工具、预算及框架分别展示。条件未说明的组保留公布值供查证。

Responses with at least one factual error %

工具条件未记录 / low / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.

Factual error rate on difficult prompts · Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified. · Responses with at least one factual error · 工具条件未记录 / low / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.
名次模型与配置成绩(%)记录与证据
1GPT-6.1 TerraGPT-6.1 Terra low16.7%
2GPT-6 LunaGPT-6 Luna low44.3%

工具条件未记录 / medium / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.

Factual error rate on difficult prompts · Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified. · Responses with at least one factual error · 工具条件未记录 / medium / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.
名次模型与配置成绩(%)记录与证据
1GPT-6.1 TerraGPT-6.1 Terra medium10.1%
2GPT-6 LunaGPT-6 Luna medium31.2%

工具条件未记录 / high / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.

Factual error rate on difficult prompts · Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified. · Responses with at least one factual error · 工具条件未记录 / high / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.
名次模型与配置成绩(%)记录与证据
1GPT-6.1 TerraGPT-6.1 Terra high7.7%
2GPT-6 LunaGPT-6 Luna high27.6%

工具条件未记录 / xhigh / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.

Factual error rate on difficult prompts · Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified. · Responses with at least one factual error · 工具条件未记录 / xhigh / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.
名次模型与配置成绩(%)记录与证据
1GPT-6.1 TerraGPT-6.1 Terra xhigh5%
2GPT-6 LunaGPT-6 Luna xhigh24.5%

工具条件未记录 / maximum / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.

Factual error rate on difficult prompts · Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified. · Responses with at least one factual error · 工具条件未记录 / maximum / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.
模型与配置成绩(%)记录与证据
GPT-6.1 TerraGPT-6.1 Terra max4.3%

工具条件未记录 / max / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.

Factual error rate on difficult prompts · Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified. · Responses with at least one factual error · 工具条件未记录 / max / 框架未记录 · Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.
模型与配置成绩(%)记录与证据
GPT-6 LunaGPT-6 Luna max18.4%

历史记录 · 5 次发布

  • GPT-6.1 Terra本次发布有 5 条成绩记录,含 5 条精确值(可含历史记录)
  • GPT-6.1 Sol本次发布有 5 条成绩记录,含 5 条精确值(可含历史记录)
  • GPT-6 Luna本次发布有 5 条成绩记录,含 5 条精确值(可含历史记录)
  • GPT-6 Astra本次发布有 5 条成绩记录,含 5 条精确值(可含历史记录)
  • GPT-5.6 Terra本次发布有 5 条成绩记录,含 5 条精确值(可含历史记录)